Rogo Narrative Analysis: 7.0 Out of 10

Verdict: Rogo's CEO told Bloomberg about a 21-year-old working at 2 a.m., while rogo.com promises to "increase efficiency."

A Narrative Authenticity Framework analysis of Rogo. NAF score 7.0 out of 10. Weakest component: Real Aspiration, 6.0.

In August 2019, a Princeton junior named Gabe Stengel was visiting a friend in the Hamptons. He ran into Peter Orszag, then a vice chairman at Lazard, and they got talking.

Stengel said his quant trading internship had left him cold. His friends' banking work interested him more. Orszag mentioned a job he was thinking of creating: a data translator who would sit between Lazard's bankers and its data scientists.

Stengel took that job after graduation. He told Puck's William Cohan he probably worked "double the hours for half of the pay."

Then he left. Fortune's Term Sheet tells the rest: he quit at 23 to build the AI that would automate his old job, sat through 30 investor passes and two hard years. Today Rogo is worth about $2 billion, and Lazard is a customer.

You won't find any of that on rogo.com. The homepage says Rogo helps institutions "increase efficiency and drive smarter decisions." The Company page names no founders at all.

Where Rogo stands

Rogo raised a $75 million Series C led by Sequoia in January 2026 at a $750 million valuation. In April, Kleiner Perkins led a $160 million Series D at about $2 billion. Total funding is now above $300 million, and nine global banks sit on its cap table.

Named clients include Lazard, Jefferies, Moelis, Rothschild & Co, Nomura, Baird and Truist. The product line runs from Felix, an agent bankers hand work to by email, to Forward Deployed Bankers, former bankers who embed with client teams. Rogo also built Big Finance Bench, a 928-question benchmark, with researchers from OpenAI.

The brand line: "Built for Finance. Rogo is the trusted AI partner helping the world's leading financial institutions increase efficiency and drive smarter decisions." The Company page states the ambition "to create the biggest financial AI company on earth."

Why the current story falls short

"Increase efficiency and drive smarter decisions" could sit on the homepage of AlphaSense, Hebbia, FactSet or a dozen bank software vendors. It names a benefit every rival claims.

The real story is sharper, and Rogo's own CEO keeps telling it to reporters. In April he told Bloomberg that much of the analytical work "is done by a 21-year-old in tools from 40 years ago at 2 a.m." That sentence has a person, a time stamp and a cost in it. The homepage has none of the three.

Bloomberg, Puck, Fortune and Contrary Research all tell the Lazard origin. Rogo links to the Bloomberg piece from its newsroom strip and never tells the story in its own voice.

The NAF scorecard

ComponentScoreOne-line read
Friction6.5The 2 a.m. problem lives in press quotes, not on the site
Lived Stake7.5A great founder story, told by everyone except Rogo
Distinctive Qualification8.0Nine bank investors and a benchmark of its own
Real Aspiration6.0Wants to be biggest; never says what changes
Composite7.0

That puts Rogo level with Gamma and Databricks on the SeekArgus leaderboard. Patagonia's 9.5 is the reference point.

Friction: 6.5

Rogo knows the pain exactly. A 2024 Rogo post said the product saved users about 10 hours a week. None of that reaches the homepage. "Increase efficiency" is category language, and the time-saved figure is two years old.

Lived Stake: 7.5

This should be Rogo's strongest score. Stengel did the job Rogo automates, at a firm that now buys the product, after a chance meeting with the man who later became Lazard's CEO. Co-founder John Willett did similar work at Barclays and JPMorgan. Forward Deployed Bankers shows the team has practitioners. The founders themselves appear nowhere on the Company page.

Distinctive Qualification: 8.0

The proof is strong: bank shareholders, a benchmark co-written with OpenAI researchers, and a client list from Lazard to Nomura. Consistency holds it at 8.0. Rogo's own channels have given three different sizes this year, the latest being 60,000 users at 450 institutions on today's homepage, and none of them carries a date.

Real Aspiration: 6.0

"The biggest financial AI company on earth" tells a banker what Rogo wants. It doesn't tell them what changes for them. That gap matters because the industry is arguing about exactly that. CNBC reported JPMorgan discussing a cut in its junior-to-senior banker ratio from 6:1 to 4:1, and McKinsey's QuantumBlack told Fortune that banks are cutting junior analyst classes by up to two-thirds.

In 2024, Stengel told Fortune that finance remains "an apprenticeship business." A 9 here would turn that line into a public promise about how the next generation of bankers learns.

The differentiated story

Lazard hired Stengel to stand between bankers and data scientists. He saw the late-night analyst work from both sides, then left to build the thing that does it. Hebbia, AlphaSense and the frontier labs can all say "AI for finance." Only Rogo can say its CEO was hired into the exact seam the product now fills, and that the firm that hired him is a client.

In 2019, Lazard created a job for Gabe Stengel: translate between its bankers and its data scientists. He spent two years watching analysts rebuild the same models at 2 a.m. Then he quit to make that work disappear. Thirty investors said no. Today 450 institutions, Lazard among them, run on what he built.

Four moves to raise the score

1. Publish what the analyst job becomes (Real Aspiration 6.0 to 7.5)

A short, signed statement from Stengel on the Company page that answers one question: what does a first-year analyst do when the 2 a.m. work is gone? Co-sign it with a client bank, and let it replace "biggest financial AI company on earth."

Prior-work check: Partial. Rogo's 2024 Series A post called finance "an apprenticeship business," and its Baird case study mentions a training curriculum. Neither describes the changed job itself. No public evidence Rogo has done this, as of November 19, 2026.

2. Tell the founder story on Rogo's own site (Lived Stake 7.5 to 8.5)

Name the three founders on the Company page and add 150 words on the origin: the Hamptons conversation, the data translator job, the double hours, the 30 passes.

Prior-work check: Partial. An earlier Company page listed the founders by name, and the current one does not. Every telling of the origin we found was written by Puck, Bloomberg, Fortune or Contrary Research. No public evidence Rogo has done this, as of November 19, 2026.

3. Name the 2 a.m. problem on the homepage (Friction 6.5 to 8.0)

Replace "increase efficiency" with the problem in Stengel's own terms, paired with one dated time-saved figure tied to a task bankers recognize, such as a comps refresh.

Prior-work check: Partial. The 2024 post claimed about 10 hours saved a week, and the Baird and Harbor View case studies carry firm-level numbers. None reaches the homepage. No public evidence Rogo has done this, as of November 19, 2026.

4. Date every number (Distinctive Qualification 8.0 to 8.5)

One "Rogo by the numbers" page with users, institutions and daily queries, each stamped with the date it was measured, and every other channel pointing to it.

Prior-work check: Not found. Big Finance Bench publishes dated scores for AI models, but Rogo publishes no dated record of its own adoption, as of November 19, 2026.

Projected composite: 8.1 (Friction 8.0, Lived Stake 8.5, Distinctive Qualification 8.5, Real Aspiration 7.5). These checks cover public sources only and can't see Rogo's internal plans.

Sources

Tim Donovan · SeekArgus

Every NAF score in this analysis is part of the SeekArgus NAF Benchmark, a proprietary dataset of 108 scored companies (as of November 2026). Scores are as of publication. How the benchmark works · Current standings
Previous
Previous

Baseten Narrative Analysis: 7.4 Out of 10

Next
Next

Tribe AI Narrative Analysis: 6.4 Out of 10